A VP-accordant checkpointing protocol preventing useless checkpoints

Baldoni, Roberto; Quaglia, Francesco; Ciciani, Bruno

doi:10.1109/reldis.1998.740475

Cited by 23 publications

(36 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Coordinated checkpointing protocols coordinate the processes at checkpoint time to ensure that the saved global state is consistent [11]. Communication-induced checkpointing provides the same guarantee without synchronizing the processes explicitly: they piggyback information on application messages instead [4]. Finally, uncoordinated checkpointing protocols take process checkpoints independently [5].…”

Section: B Modeling Applications Determinismmentioning

confidence: 99%

HydEE: Failure Containment without Event Logging for Large Scale Send-Deterministic MPI Applications

Guermouche

Ropars

Snir

et al. 2012

2012 IEEE 26th International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

Abstract-High performance computing will probably reach exascale in this decade. At this scale, mean time between failures is expected to be a few hours. Existing fault tolerant protocols for message passing applications will not be efficient anymore since they either require a global restart after a failure (checkpointing protocols) or result in huge memory occupation (message logging). Hybrid fault tolerant protocols overcome these limits by dividing applications processes into clusters and applying a different protocol within and between clusters. Combining coordinated checkpointing inside the clusters and message logging for the inter-cluster messages allows confining the consequences of a failure to a single cluster, while logging only a subset of the messages. However, in existing hybrid protocols, event logging is required for all application messages to ensure a correct execution after a failure. This can significantly impair failure free performance. In this paper, we propose HydEE, a hybrid rollback-recovery protocol for send-deterministic message passing applications, that provides failure containment without logging any event, and only a subset of the application messages. We prove that HydEE can handle multiple concurrent failures by relying on the senddeterministic execution model. Experimental evaluations of our implementation of HydEE in the MPICH2 library show that it introduces almost no overhead on failure free execution.

show abstract

Section: B Modeling Applications Determinismmentioning

confidence: 99%

HydEE: Failure Containment without Event Logging for Large Scale Send-Deterministic MPI Applications

Guermouche

Ropars

Snir

et al. 2012

2012 IEEE 26th International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

show abstract

“…Because ckpt_index(I q,last ) = CI max , q rolls back to the last checkpoint, C q,7 . Because there is no checkpoint interval I such that I r,3 depends on I and the checkpoint at the end of I is not taken, process r can roll back to C r, 8 . Likewise, process p and s roll back to their last check-…”

Section: Rulementioning

confidence: 99%

“…Asynchronous checkpointing [3,4] and quasi-synchronous checkpointing [5][6][7][8][9][10][11][12] protocols allow processes to take checkpoints at different frequencies. Because asynchronous checkpointing protocols do not ensure the advance of consistent global checkpoint, the protocols are susceptible to the domino effect [13].…”

Section: Introductionmentioning

confidence: 99%

A multi-cycle checkpointing protocol that ensures strict 1-rollback

Zhang

Zuo

et al. 2012

Information Processing Letters

View full text Add to dashboard Cite

“…However, log-based protocols are predominantly used in databases [28] and file systems [34] because they have access to a large and persistent disk storage. CRR is checkpointbased, and there is an extensive set of algorithms and implementations of distributed checkpoint schemes in a variety of domains ranging from loosely coupled message passing systems to tightly coupled multiprocessors [5,6,10,19,22,30,32,36,38]. A taxonomy and survey of such schemes is given in [20].…”

Section: Related Workmentioning

confidence: 99%

Declarative failure recovery for sensor networks

Gummadi

Kothari

Millstein

et al. 2007

Proceedings of the 6th International Conference on Aspect-Oriented Software Development

View full text Add to dashboard Cite

Wireless sensor networks consist of a system of distributed sensors embedded in the physical world, and promise to allow observation of previously unobservable phenomena. Since they are exposed to unpredictable environments, sensor-network applications must handle a wide variety of faults: software errors, node and link failures, and network partitions. The code to manually detect and recover from faults crosscuts the entire application, is tedious to implement correctly and efficiently, and is fragile in the face of program modifications. We investigate language support for modularly managing faults. Our insight is that such support can be naturally provided as an extension to existing "macroprogramming" systems for sensor networks. In such a system, a programmer describes a sensor network application as a centralized program; a compiler then produces equivalent node-level programs. We describe a simple checkpoint API for macroprograms, which can be automatically implemented in a distributed fashion across the network. We also describe declarative annotations that allow programmers to specify checkpointing strategies at a higher level of abstraction. We have implemented our approach in the Kairos macroprogramming system. Experiments show it to improve application availability by an order of magnitude and incur low messaging overhead.

show abstract

A VP-accordant checkpointing protocol preventing useless checkpoints

Cited by 23 publications

References 14 publications

HydEE: Failure Containment without Event Logging for Large Scale Send-Deterministic MPI Applications

HydEE: Failure Containment without Event Logging for Large Scale Send-Deterministic MPI Applications

A multi-cycle checkpointing protocol that ensures strict 1-rollback

Declarative failure recovery for sensor networks

Contact Info

Product

Resources

About